Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Ratheesh Kannoth <rkannoth@marvell.com>
To:     <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>
CC:     <sgoutham@marvell.com>, <davem@davemloft.net>,
        <edumazet@google.com>, <kuba@kernel.org>, <pabeni@redhat.com>,
        <sbhatta@marvell.com>, <gakula@marvell.com>, <schalla@marvell.com>,
        <hkelam@marvell.com>, "Ratheesh Kannoth" <rkannoth@marvell.com>
Subject: [PATCH net-next] octeontx2-pf: TC flower offload support for rxqueue mapping
Date:   Mon, 19 Jun 2023 11:36:38 +0530
Message-ID: <20230619060638.1032304-1-rkannoth@marvell.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Type: text/plain
Precedence: bulk

 TC rule support to offload rx queue mapping rules.

Eg:
   tc filter add dev eth2 ingress protocol ip flower \
      dst_ip 192.168.8.100  \
      action skbedit queue_mapping 4 skip_sw
      action mirred ingress redirect dev eth5

Packets destined to 192.168.8.100 will be forwarded to rx
queue 4 of eth5 interface.

   tc filter add dev eth2 ingress protocol ip flower \
      dst_ip 192.168.8.100  \
      action skbedit queue_mapping 9 skip_sw

Packets destined to 192.168.8.100 will be forwarded to rx
queue 4 of eth2 interface.

Signed-off-by: Ratheesh Kannoth <rkannoth@marvell.com>
---
 .../net/ethernet/marvell/octeontx2/nic/otx2_tc.c   | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c
index 231c3f0efb60..8a13df592af6 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c
@@ -396,8 +396,12 @@ static int otx2_tc_parse_actions(struct otx2_nic *nic,
 				return -EOPNOTSUPP;
 			}
 			req->vf = priv->pcifunc & RVU_PFVF_FUNC_MASK;
-			req->op = NIX_RX_ACTION_DEFAULT;
-			return 0;
+
+			/* if op is already set; avoid overwriting the same */
+			if (!req->op)
+				req->op = NIX_RX_ACTION_DEFAULT;
+			break;
+
 		case FLOW_ACTION_VLAN_POP:
 			req->vtag0_valid = true;
 			/* use RX_VTAG_TYPE7 which is initialized to strip vlan tag */
@@ -433,6 +437,12 @@ static int otx2_tc_parse_actions(struct otx2_nic *nic,
 		case FLOW_ACTION_MARK:
 			mark = act->mark;
 			break;
+
+		case FLOW_ACTION_RX_QUEUE_MAPPING:
+			req->op = NIX_RX_ACTIONOP_UCAST;
+			req->index = act->rx_queue;
+			break;
+
 		default:
 			return -EOPNOTSUPP;
 		}
-- 
2.25.1